You are talking about Program Options library. I don't quite agree it covers the same problem domain. I feel it is more focued on command-line than on reading configuration in general. The biggest difference is that property_tree is a DOM, not a linear structure.
If that's DOM, then why the interface is so radically different from W3W DOM spec? I understand that you mean "DOM" in the broad sense, but anyway, all libraries that pretend to working with XML are using specific DOM interface. It might not be excellent, but it allows one to write essentially the same code using all existing libs.
The reason why it does not use W3C DOM standard, is because this is not an XML library. It is meant to support as much hierarchical file formats as possible. I believe this can be achieved only through making it as general as possible - thus only one string per data and one string per key (apart from traits customization, which is another story). No "special" nodes, like comments, leaves/branches, attributes etc. The nice side effect of that simplicity is that the library is extremely easy to use in many cases. Think about _three_ lines of code and one include to read arbitrarily-typed value from a config file - in any supported format. I don't think this can be matched by W3C DOM. Best regards, Marcin